The sputum in influenza bronchitis varies. "Characteristic influenza sputum" is frequently mentioned. But there are various kinds. Especially characteristic is the copious sputum which is often expectorated in remarkable quantities, which we frequently found collected in several vessels on making our hospital round. In these cases a large layer of white or dirty gray foam lies on top; next follows a layer composed of a quantity of turbid serous material in which purulent flakes are suspended, forming the principal portion, while the bottom of the receptacle is covered by a thin layer of vitreous like mucus or by a turbid deposit. Another highly characteristic variety of sputum is found early at the height of the influenza attack; it is an almost purulent, nummular, occasionally globular sputum, in the latter case quite like the cavernous expectoration of tubercular individuals. The "yellowish green nummular sputum" is considered by R. Pfeiffer as "a characteristic peculiarity of influenza." There is yet a third variety of sputum to be mentioned as frequently occurring in influenza bronchitis, in which bloody streaks or pure blood of dark red color, either nummular or globular in shape, is mixed with the purulent or serous expectoration, reminding one of the sputum in infarction of the lung. This is due to a rupture of vessels in the hyperemic bronchial mucous membrane, either spontaneous or caused by the exertion of coughing (bronchitis hemorrhagica grippalis). The bronchitic hemoptysis in influenza was mentioned already by Nelson (1803), Leared (1862), and others.

If the influenza bronchitis extends diffusely into the bronchioles, there arise dyspnea and cyanosis. Old decrepit individuals, tuberculous patients, and weak, rachitic children may succumb to this diffuse capillary bronchitis at the height of an influenza attack without any associated pneumonia.

An almost unnoticed fact is the occurrence of an acute bronchiectasis of the middle and smaller bronchi during the influenza attack. We have seen many instances of this condition at postmortem examinations. This possibility must also be kept in mind during life, if, after an influenza has run its course, there should persist, over one or more pulmonary lobules with a normal percussion note, numerous large vesicular metallic rales and copious sputum of a bron chiectatic character. Such cases of bronchial dilatation may last weeks and even months and yet terminate in complete recovery. This was the case in one of our patients in whom, on account of their long duration, we had thought that the bronchiectatic cavities and interstitial changes would be permanent. In consequence of the bronchial dilatation a fetid, putrid bronchitis may also arise.

The occurrence of a fibrinous or croupous bronchitis with or without pneumonia has been noted in the recent epidemics by Naunyn and Weichselbaum, and by Nonat, Magendie, Caseaux, Legendre, in the earlier epidemics.

It will be appropriate to describe here the anatomic picture of influenza bronchitis. It is characterized by an exceptionally intense hyperemia of the mucous membranes, extending into the finest bronchi. On pressure the lung exudes, on to the dark red, cut surface, numerous large and small tenacious, whitish yellow drops of pus from the severed bronchi. Kundrat correctly states: "These bronchitides were characterized both by their extent and by their intensity, inasmuch as not only the bronchi of the posterior but also those of the anterior and superior portions of the lung were found occluded. Not only the finer, but also the middle sized, bronchi were occluded by thick, mucopurulent secretion, clumpy in the larger bronchi, and giving the impression that the individuals had been suffocated by their bronchial secretions." This is the "catarrhe asphyxique grippal" of Netter.

We shall give only a brief description of the finer histologic changes which influenza produces in the respiratory mucous membrane. They have been described by the pathologists: Ribbert, Klebs, Kuskow, Mar chand, and others.

The chief point is the enormous hyperemia, which not rarely gives rise to extravasation of blood. Attention is also called to the profuse cellular infiltration, which, according to Kuskow, often gives rise to the formation of "small lymphomata." Furthermore, the [frequency and significance of thrombosis are commented upon, especially by Klebs and Kuskow. Klebs mentions only " capillary throm-'boses," while Kuskow considers thromboses of the smaller arteries ',and veins more common. The thrombi consist of small, finely granular, fibrinous masses, which gradually assume a more homogeneous character until they appear hyaline. Micro organisms occasionally, although rarely, give rise to the formation of numerous obstructing thrombi. Kuskow also describes hypertrophy and exfoliation of the vascular endothelium and the resulting "desquamation thrombi."

Obstruction of the larger capillary areas causes necrosis of the mucous membrane, which is generally confined to the superficial layers; nevertheless, on one occasion Kuskow found a necrosis extending deep into the submucosa, with exposure of the cartilages of the trachea and of the larger bronchi, together with an ulcerative tracheitis and bronchitis analogous to the ulcerative laryngitis previously mentioned.

The processes affecting the tracheobronchial mucous membrane, already described,-hyperemia, small celled infiltration, thrombus formation,-occur in a similar manner and even in a more marked degree in the lungs. The disposition of inflammatory pulmonary processes occurring during influenza to abscess and gangrene formation is thereby easily explained.

The conception arrived at by Klebs, which was based upon an anatomic investigation, is worthy of mention. In his opinion the processes taking place in the blood constitute the primary and most important phenomena in influenza. The apparently primary catarrhal and inflammatory manifestations, which, according to the opinion of many, represent the essential points of the process, are, according to Klebs, nothing else than the results of the obstruction of large capillary areas by thrombi. These are formed by granular masses in the blood derived from the leukocytes and erythrocytes. From the resolution of these soft granular masses are formed masses of thrombi becoming more and more homogeneous. The obstruction of larger capillary areas causes hyperemia of the neighboring capillaries, inflammation, and exudation, and favors the deposition of pathogenic micro organisms, which find a suitable soil in the tissues injured by the thrombosis.

Klebs would also attribute the cerebral manifestations and other functional affections of the nervous system, as psychoses, somnolence, and coma, to circulatory disturbances, which in their turn have arisen from capillary thromboses. He believes that, as compared with these anatomic phenomena, the toxic influences play but a small part. Our clinical experience does not agree with the hypotheses of Klebs.